The migration of vascular smooth muscle cells (VSMCs) from the media to the intima and the remodelling of the vascular extracellular matrix (ECM) are interrelated key pathogenic features of a number of life-threatening vascular conditions, such as atherosclerosis and restenosis following balloon angioplasty. Remodeling of the vascular ECM also occurs with aging and involves changes that are likely to contribute to the increased incidence and severity of vessel disease in aging animal populations. VSMCs in vivo are surrounded by and embedded in ECMs that must be traversed during migration, a process that involves regulated extracellular proteolysis. We have previously shown that activation of matrix metalloproteases (MMPs), particularly that of MMP-2 or type IV collagenase, occurs is required for the in vitro migration of VSMCs across an ECM barrier. We now demonstrate that activation of MMP2 occurs following balloon injury and that the activator of MMP2, known as membrane type-MMP(MT-MMP), is upregulated in vivo in response to vessel injury and in vitro in response to ligand engagement of a2b1 integrin. Given that expression of a2b1 is increased in VSMCs in vivo following vessel injury, signaling through this integrin may link injury to activation of the MMP cascade. In addition to acting as a barrrier for cell movement, the ECM may also regulate VSMC differentiation. The molecular mechanisms controlling VSMC differentiation are unknown but may involve the family of ID proteins (for inhibitor of DNA binding or differentiation). ID proteins are helix-loop-helix proteins that act in many in other cell systems as transdominant suppressors of differentiation. We show that many ID isoforms, including a unique splice variant of ID3, are expressed in dedifferentiated VSMCs. Forced expression of, at least, one of thses IDs, namely rat ID1, blocks VSMC differentation. Current studies are examining the effects of the splice variant of ID3 which we have cloned from proliferating VSMCs. The 2nd helix of the HLH domain, a region critically important for the function of ID, is disrupted in this variant. We have shown that this variant does not block the transactivating action of the basic-HLH protein, myoD, in 10T1/2 or the growth-suppressing effects of the retinoblastoma gene product in SAOS cells. In fact, the splice variant of ID3 acts as a dominant inhibitor of ID3 in SAOS proliferation suppresion assay.